General Visualizations
This section presents visualizations from the study “Informal
Employment and Business Cycles: Stylized Facts for Argentina
(2004–2024)”, developed by Pedro P. Cohan and Francisco Leiva.
All visualizations were produced by the authors, based on the technical
tables of the
Permanent Household Survey (Instituto Nacional de Estadística y Censos,
INDEC) .
Table 1 | Occupational categories. Total agglomerates of the EPH. Average relative proportion from quarterly data in a three-year environment.
Figure 1 | Population and activity and inactivity levels. Number of inhabitants represented in the nationwide
Note: EAP: economic active population = force labor.
Figure 2 | Formal and informal workers. Number of inhabitants represented in the survey with national coverage. Series filtered by seasonality.
Figure 3a | Distribution of quarterly variations of formal employment. Period 2004.Q1–2024.Q4 (excluding 2020.Q2–2021.Q2).
Figure 3b | Distribution of quarterly variations of informal employment. Period 2004.Q1–2024.Q4 (excluding 2020.Q2–2021.Q2).
Figure 4 | Volatility of each variable relative to the amplitude of the Argentine business cycle.
Table 2 | Correlation coefficients between the quarterly variations of each variable with respect to the variations of the ICA-ARG.
Seasonally adjusted data. Summary of results for +/- four quarters lead and lag. Period 2004.Q1-2024.Q4 (excluding 2020.Q2-2021.Q2)
Figure 5 | Percent deviations from trend for each variable considered.
Table 3 | Regression results for each variable on activity
Estimated model: \(PDT_i = \beta_i \,
PDT_{(ICA\text{-}ARG)} + \varepsilon_i\),
based on the percent deviation from trend of each variable, using
seasonally adjusted data.
Notes on regressions
- Regression I: Entire period (2004.Q1–2024.Q4, n = 81)
- Regression II: Pre-pandemic period (2004.Q1–2020.Q1, n = 62)
- Regression III: Entire period (2021.Q2–2024.Q4, n = 19) [Not showed on presented paper]
- Regression IV: Entire period (2004.Q1–2020.Q1 & 2021.Q2–2024.Q4, n = 76) [Not showed on presented paper]
Table 4 | Elasticities of total employment to changes in occupational categories
Notes on regression
\[ \Delta \log(y_t) = \log(y_t) - \log(y_{t-1}) \]
In an extended form, the exercise involved regressing the total number of persons employed (dependent variable) against the formal and informal components of each occupational category (explanatory variables), which enabled the analysis of the elasticities of the level of employment with respect to its components:
\[ \begin{aligned} \Delta \log(\text{employment}) =\;& \beta_{1}\,\Delta \log(\text{salaried workers formal}) + \beta_{2}\,\Delta \log(\text{salaried workers informal}) \\[6pt] &+ \beta_{3}\,\Delta \log(\text{self-employed workers formal}) + \beta_{4}\,\Delta \log(\text{self-employed workers informal}) \\[6pt] &+ \beta_{5}\,\Delta \log(\text{employers formal}) + \beta_{6}\,\Delta \log(\text{employers informal}) \\[6pt] &+ \beta_{7}\,\Delta \log(\text{family workers informal}) + \varepsilon \end{aligned} \]
Figure 6 | Number of persons listed by occupational category. All agglomerates in the EPH. Series filtered by seasonality. Quarterly data: period 2004.Q1–2024.Q4.